Security, Privacy & Safety
AI Safety Engineer, Medical Deployment
New York, NYFull-time
Focus on failure modes that matter in healthcare: unsafe recommendations, silent degradations, unreliable escalation, and misuse. You will help create the safety operating layer between advanced models and real patient outcomes.
Responsibilities
- •Build safeguards, release criteria, and runtime checks for medical model behavior
- •Investigate failure cases across product, operations, and clinical review workflows
- •Work with research to define practical mitigation strategies for unsafe or low-confidence outputs
- •Partner with security and clinical teams on escalation design and red-teaming
- •Help define what responsible deployment means for medical AI in the real world
Requirements
- •Strong technical background in ML systems, safety engineering, or high-reliability software
- •Experience evaluating or mitigating failure modes in model-powered systems
- •Ability to reason about operational risk, product behavior, and human factors together
- •Comfort working across research, engineering, and policy-adjacent questions
- •Healthcare or safety-critical system experience is strongly preferred